summaryrefslogtreecommitdiffstats
path: root/problem.tex
diff options
context:
space:
mode:
authorStratis Ioannidis <stratis@stratis-Latitude-E6320.(none)>2013-07-04 15:24:11 -0700
committerStratis Ioannidis <stratis@stratis-Latitude-E6320.(none)>2013-07-04 15:24:11 -0700
commit0f9dd86764035eec8a8f79cd0a457d08d2522dae (patch)
treefde88fb5d1bad7786ee13ebe1f16455dd1836704 /problem.tex
parenta53c28eebf8926d02a835dc55a903b6528c8e784 (diff)
downloadrecommendation-0f9dd86764035eec8a8f79cd0a457d08d2522dae.tar.gz
mechanism
Diffstat (limited to 'problem.tex')
-rw-r--r--problem.tex4
1 files changed, 2 insertions, 2 deletions
diff --git a/problem.tex b/problem.tex
index a00e2f6..6f07d29 100644
--- a/problem.tex
+++ b/problem.tex
@@ -79,12 +79,12 @@ d}.$ Intuitively, this corresponds to the simplest prior, in which no direction
of $\reals^d$ is a priori favored; equivalently, it also corresponds to the
case where ridge regression estimation \eqref{ridge} performed by $\E$ has
a penalty term $\norm{\beta}_2^2$. A generalization of our results to arbitrary
-covariance matrices $R$ can be found in Section~\ref{sec:ext}.
+covariance matrices $R$ can be found in Appendix~\ref{sec:ext}.
%Note that \eqref{dcrit} is a submodular set function, \emph{i.e.},
%$V(S)+V(T)\geq V(S\cup T)+V(S\cap T)$ for all $S,T\subseteq \mathcal{N}$; it is also monotone, \emph{i.e.}, $V(S)\leq V(T)$ for all $S\subset T$.
-\subsection{Budget-Feasible Experimental Design: Full Information Case}
+\subsection{Budget-Feasible Experimental Design: Full Information Case}\label{sec:fullinfo}
Beyond the cardinality constraint in classical experimental design discussed above, a
budgeted version can also be considered.